Abstract: We present COD3S, a novel method for generating semantically diverse sentences using neural sequence-to-sequence (seq2seq) models. Conditioned on an input, seq2seqs typically produce semantically and syntactically homogeneous sets of sentences and thus perform poorly on one-to-many sequence generation tasks. Our two-stage approach improves output diversity by conditioning generation on locality-sensitive hash (LSH)-based semantic sentence codes whose Hamming distances highly correlate with human judgments of semantic textual similarity. Though it is generally applicable, we apply \method{} to causal generation, the task of predicting a proposition's plausible causes or effects. We demonstrate through automatic and human evaluation that responses produced using our method exhibit improved diversity without degrading task performance.